What makes the "Rule of Thirds" work?

"TheDave©" wrote in message
...
Yeah, yeah, I know, rules are made to be broken, be creative, think
outside the box, it doesn't always work, for every rule there's an
exception, yada yada yada... but, it's still a much-taught and accepted
"rule".

What makes it work?
The design of the human brain.
Jim

"TheDave©" wrote in message
...
Yeah, yeah, I know, rules are made to be broken, be creative, think
outside the box, it doesn't always work, for every rule there's an
exception, yada yada yada... but, it's still a much-taught and
accepted "rule".

What makes it work?


Well, it doesn't work all that brilliantly as a 'universal' rule. That is,
while the Golden section works in a rectangle of any proportion, or a
square, thirds really doesn't.

Thirds works OK for a square, but where it works best is in a format with
the proportions of the 35mm frame, which are 24:36 or 2:3. This natural
threes relationship in the frame makes the 'thirds' seem pleasing to the
eye. If you routinely print to a different proportion, you'll find that the
'rule' of thirds doesn't 'work' nearly so well any more once that 2:3
proportion of the frame no longer applies.

Interestingly, it doesn't look as good in the 6x17 format either. Maybe
with the 2:3 proportions of 35mm the relationship is just subtle enough for
the brain to enjoy it subconsciously. When you move to the 1:3 proportions
of 617 (yes, I know, but the _actual_ frame size is 1:3) then it becomes too
obvious, the conscious mind 'sees' it and the compostion can look mannered
at best, trite at worst. Well, that's how it frequently seems to me,
anyway.

Thus the 'rule' of thirds is more a rule about matching your composition to
the proportions of your frame/format - which is what we should be trying to
do visually anyway.



Peter

"Tony Polson" wrote in message
...
Paul Furman wrote:

Tony Polson wrote:

The "Rule of Thirds" is actually a very inaccurate
approximation of the "Golden Ratio" or "Golden Section",
which has a long historic basis in composition of landscape
paintings.

The Golden Ratio is not as simple as dividing the picture into
thirds. It is based on dividing one side of the composition so
that the ratio of the smaller part to the larger part is the same
as the ratio of the larger part to the whole.

Hmm so se need a Golden Section focusing screen grid. And a
button to flip, rotate & mirror it :-)


I already have one. ;-)


We all do - it's wired into our brains... :-)


Peter

"Colin_D" wrote in message
...
TheDave© wrote:
[SNIP]

Practically, there isn't a lot of difference between thirds and the
Golden Mean/Section. A point on the thirds is 33.3% in from
adjacent edges of an image, while a Golden Mean/Section is
38.2% in from adjacent edges, slightly closer to the centre of the
image. Given that a picture element placed on thirds or GS is
probably considerably larger than the difference between the
point positions, the 4.9% difference is likely to be indiscernible.


Try it and see: I find that it's not indiscernible at all to my eye.

It's interesting to analyse a number of pictures that different people have
composed by eye - ie. they shot what they thought looked best, not what
conformed to some rule. If you do this, you'll find a (perhaps
surprisingly) large number of the best ones hit the Golden Section with
their key picture elements - especially those pictures with a feeling of
stillness or calm to them.

(That can be the biggest criticism of composing that way deliberately: it
produces a picture that feels harmonious even when maybe it shouldn't, so
the 'rule' is not always appropriate - like all such 'rules'.)

If you do this exercise with pictures of different proportions you'll also
find fewer and fewer pictures where the thirds 'works' in this form of
intutitive rather than rule driven composition the further you move away
from the 2:3 frame proportions of 35mm, whereas the Goldn Section continues
to 'work'.


Peter

"Bandicoot" wrote:

"Tony Polson" wrote in message
.. .
Paul Furman wrote:

Tony Polson wrote:

The "Rule of Thirds" is actually a very inaccurate
approximation of the "Golden Ratio" or "Golden Section",
which has a long historic basis in composition of landscape
paintings.

The Golden Ratio is not as simple as dividing the picture into
thirds. It is based on dividing one side of the composition so
that the ratio of the smaller part to the larger part is the same
as the ratio of the larger part to the whole.

Hmm so se need a Golden Section focusing screen grid. And a
button to flip, rotate & mirror it :-)


I already have one. ;-)


We all do - it's wired into our brains... :-)


That is so true!

But I doubt many people here would even know what you meant.

;-)

"Bandicoot" wrote:

"Colin_D" wrote in message
.. .
TheDave© wrote:
[SNIP]

Practically, there isn't a lot of difference between thirds and the
Golden Mean/Section. A point on the thirds is 33.3% in from
adjacent edges of an image, while a Golden Mean/Section is
38.2% in from adjacent edges, slightly closer to the centre of the
image. Given that a picture element placed on thirds or GS is
probably considerably larger than the difference between the
point positions, the 4.9% difference is likely to be indiscernible.


Try it and see: I find that it's not indiscernible at all to my eye.


Absolutely right. It is not only discernible, it is a significant
difference.

Looking at the ratios will tell you why, because the Golden Section is
all about the ratios being the same at 1.618, whereas the ratios on
the Rule of Thirds are very different - 1.500 against 2.000.

I suppose 1.500 and 2.000 have a greater appeal to simpler people. It
takes a little sophistication to understand why 1.618 works and the
Rule of Thirds does not.

;-)

Tony Polson wrote:
"Bandicoot" wrote:

"Colin_D" wrote in message
...
TheDave© wrote:
[SNIP]
Practically, there isn't a lot of difference between thirds and the
Golden Mean/Section. A point on the thirds is 33.3% in from
adjacent edges of an image, while a Golden Mean/Section is
38.2% in from adjacent edges, slightly closer to the centre of the
image. Given that a picture element placed on thirds or GS is
probably considerably larger than the difference between the
point positions, the 4.9% difference is likely to be indiscernible.

Try it and see: I find that it's not indiscernible at all to my eye.


Absolutely right. It is not only discernible, it is a significant
difference.

On a 10x8 print, the Golden Section position is 3.81 inches up from the
bottom edge, and the thirds position is 3.33 inches up, a difference of
0.48 inches, or 4.8% of the 10-inch dimension, and 0.38 inches in the
8-inch dimension. Given a reasonable sized object at either of those
points, say a building or a tree, the less than ½-inch difference will
not be noticeable.

Looking at the ratios will tell you why, because the Golden Section is
all about the ratios being the same at 1.618, whereas the ratios on
the Rule of Thirds are very different - 1.500 against 2.000.

1.5 to 2 (no need for the trailing zeroes) is 1:4 or 3:4, that is
25%/75%. No way is that thirds. Do you mean 1 to 1.5, i.e. 1:3 or 2:3,
33.33%/66.67%?

I suppose 1.500 and 2.000 have a greater appeal to simpler people. It
takes a little sophistication to understand why 1.618 works and the
Rule of Thirds does not.

Yes, they would have to be simple to mistake 1.5 to 2 as a third - or
two-thirds. {:-)

Colin D.
;-)


--
Posted via a free Usenet account from http://www.teranews.com

Tony Polson wrote:
"Bandicoot" wrote:


"Colin_D" wrote in message
. ..

TheDave© wrote:

[SNIP]

Practically, there isn't a lot of difference between thirds and the
Golden Mean/Section. A point on the thirds is 33.3% in from
adjacent edges of an image, while a Golden Mean/Section is
38.2% in from adjacent edges, slightly closer to the centre of the
image. Given that a picture element placed on thirds or GS is
probably considerably larger than the difference between the
point positions, the 4.9% difference is likely to be indiscernible.


Try it and see: I find that it's not indiscernible at all to my eye.



Absolutely right. It is not only discernible, it is a significant
difference.

Looking at the ratios will tell you why, because the Golden Section is
all about the ratios being the same at 1.618, whereas the ratios on
the Rule of Thirds are very different - 1.500 against 2.000.

Er, not quite. 3/2. And you used to say 1:2. Please be consistent,
even if wrong.


I suppose 1.500 and 2.000 have a greater appeal to simpler people. It
takes a little sophistication to understand why 1.618 works and the
Rule of Thirds does not.

As very few elements in a scene are "points" but subjects, it is rare
indeed that they can map to precisely to 1.618.

It's not about sophistication, certainly in any of your photos that
we've seen.

--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: Remove FreeLunch.

Pudentame wrote:


In a vague way, I remember reading that it has something to do with the
mathematics that describe how the universe is "constructed".

The golden ratio (mean, number, whatever) comes up in nature in a sense
of "efficency". An oft cited example being the packing of seeds in the
middle of a flower. They very often follow the Fibonacci suite
(1,1,2,3,5,8,13,21 ...) per ring working outwords. Successive ratios of
the suite are very close to the Golden Ratio (esp as the the numbers get
larger).

Cheers,
Alan.

--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: Remove FreeLunch.

Tony Polson wrote:
But I doubt many people here would even know what you meant.

Least of all you. But please prove me wrong and post some images that
show your natural or deliberate famillarity with the golden ratio.

--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: Remove FreeLunch.